From: John Garry <john.garry@huawei.com> To: <damien.lemoal@opensource.wdc.com>, <joro@8bytes.org>, <will@kernel.org>, <jejb@linux.ibm.com>, <martin.petersen@oracle.com>, <hch@lst.de>, <m.szyprowski@samsung.com>, <robin.murphy@arm.com> Cc: <linux-doc@vger.kernel.org>, <linux-kernel@vger.kernel.org>, <linux-ide@vger.kernel.org>, <iommu@lists.linux-foundation.org>, <linux-scsi@vger.kernel.org>, <liyihang6@hisilicon.com>, <chenxiang66@hisilicon.com>, <thunder.leizhen@huawei.com>, John Garry <john.garry@huawei.com> Subject: [PATCH 3/4] scsi: core: Cap shost max_sectors according to DMA optimum mapping limits Date: Fri, 20 May 2022 16:23:22 +0800 [thread overview] Message-ID: <1653035003-70312-4-git-send-email-john.garry@huawei.com> (raw) In-Reply-To: <1653035003-70312-1-git-send-email-john.garry@huawei.com> Streaming DMA mappings may be considerably slower when mappings go through an IOMMU and the total mapping length is somewhat long. This is because the IOMMU IOVA code allocates and free an IOVA for each mapping, which may affect performance. For performance reasons set the request_queue max_sectors from dma_opt_mapping_size(), which knows this mapping limit. In addition, the shost->max_sectors is repeatedly set for each sdev in __scsi_init_queue(). This is unnecessary, so set once when adding the host. Signed-off-by: John Garry <john.garry@huawei.com> --- drivers/scsi/hosts.c | 5 +++++ drivers/scsi/scsi_lib.c | 4 ---- 2 files changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c index f69b77cbf538..a3ae6345473b 100644 --- a/drivers/scsi/hosts.c +++ b/drivers/scsi/hosts.c @@ -225,6 +225,11 @@ int scsi_add_host_with_dma(struct Scsi_Host *shost, struct device *dev, shost->cmd_per_lun = min_t(int, shost->cmd_per_lun, shost->can_queue); + if (dma_dev->dma_mask) { + shost->max_sectors = min_t(unsigned int, shost->max_sectors, + dma_opt_mapping_size(dma_dev) >> SECTOR_SHIFT); + } + error = scsi_init_sense_cache(shost); if (error) goto fail; diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index 8d18cc7e510e..2d43bb8799bd 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -1884,10 +1884,6 @@ void __scsi_init_queue(struct Scsi_Host *shost, struct request_queue *q) blk_queue_max_integrity_segments(q, shost->sg_prot_tablesize); } - if (dev->dma_mask) { - shost->max_sectors = min_t(unsigned int, shost->max_sectors, - dma_max_mapping_size(dev) >> SECTOR_SHIFT); - } blk_queue_max_hw_sectors(q, shost->max_sectors); blk_queue_segment_boundary(q, shost->dma_boundary); dma_set_seg_boundary(dev, shost->dma_boundary); -- 2.26.2
WARNING: multiple messages have this Message-ID (diff)
From: John Garry via iommu <iommu@lists.linux-foundation.org> To: <damien.lemoal@opensource.wdc.com>, <joro@8bytes.org>, <will@kernel.org>, <jejb@linux.ibm.com>, <martin.petersen@oracle.com>, <hch@lst.de>, <m.szyprowski@samsung.com>, <robin.murphy@arm.com> Cc: linux-scsi@vger.kernel.org, linux-doc@vger.kernel.org, liyihang6@hisilicon.com, linux-kernel@vger.kernel.org, linux-ide@vger.kernel.org, iommu@lists.linux-foundation.org Subject: [PATCH 3/4] scsi: core: Cap shost max_sectors according to DMA optimum mapping limits Date: Fri, 20 May 2022 16:23:22 +0800 [thread overview] Message-ID: <1653035003-70312-4-git-send-email-john.garry@huawei.com> (raw) In-Reply-To: <1653035003-70312-1-git-send-email-john.garry@huawei.com> Streaming DMA mappings may be considerably slower when mappings go through an IOMMU and the total mapping length is somewhat long. This is because the IOMMU IOVA code allocates and free an IOVA for each mapping, which may affect performance. For performance reasons set the request_queue max_sectors from dma_opt_mapping_size(), which knows this mapping limit. In addition, the shost->max_sectors is repeatedly set for each sdev in __scsi_init_queue(). This is unnecessary, so set once when adding the host. Signed-off-by: John Garry <john.garry@huawei.com> --- drivers/scsi/hosts.c | 5 +++++ drivers/scsi/scsi_lib.c | 4 ---- 2 files changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c index f69b77cbf538..a3ae6345473b 100644 --- a/drivers/scsi/hosts.c +++ b/drivers/scsi/hosts.c @@ -225,6 +225,11 @@ int scsi_add_host_with_dma(struct Scsi_Host *shost, struct device *dev, shost->cmd_per_lun = min_t(int, shost->cmd_per_lun, shost->can_queue); + if (dma_dev->dma_mask) { + shost->max_sectors = min_t(unsigned int, shost->max_sectors, + dma_opt_mapping_size(dma_dev) >> SECTOR_SHIFT); + } + error = scsi_init_sense_cache(shost); if (error) goto fail; diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index 8d18cc7e510e..2d43bb8799bd 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -1884,10 +1884,6 @@ void __scsi_init_queue(struct Scsi_Host *shost, struct request_queue *q) blk_queue_max_integrity_segments(q, shost->sg_prot_tablesize); } - if (dev->dma_mask) { - shost->max_sectors = min_t(unsigned int, shost->max_sectors, - dma_max_mapping_size(dev) >> SECTOR_SHIFT); - } blk_queue_max_hw_sectors(q, shost->max_sectors); blk_queue_segment_boundary(q, shost->dma_boundary); dma_set_seg_boundary(dev, shost->dma_boundary); -- 2.26.2 _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu
next prev parent reply other threads:[~2022-05-20 8:29 UTC|newest] Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top 2022-05-20 8:23 [PATCH 0/4] DMA mapping changes for SCSI core John Garry 2022-05-20 8:23 ` John Garry via iommu 2022-05-20 8:23 ` [PATCH 1/4] dma-mapping: Add dma_opt_mapping_size() John Garry 2022-05-20 8:23 ` John Garry via iommu 2022-05-20 23:32 ` Damien Le Moal 2022-05-20 23:32 ` Damien Le Moal via iommu 2022-05-20 8:23 ` [PATCH 2/4] dma-iommu: Add iommu_dma_opt_mapping_size() John Garry 2022-05-20 8:23 ` John Garry via iommu 2022-05-20 23:33 ` Damien Le Moal 2022-05-20 23:33 ` Damien Le Moal via iommu 2022-05-23 7:01 ` John Garry via iommu 2022-05-23 7:01 ` John Garry 2022-05-23 7:32 ` Damien Le Moal 2022-05-23 7:32 ` Damien Le Moal via iommu 2022-05-23 7:34 ` Damien Le Moal via iommu 2022-05-23 7:34 ` Damien Le Moal 2022-05-20 8:23 ` John Garry [this message] 2022-05-20 8:23 ` [PATCH 3/4] scsi: core: Cap shost max_sectors according to DMA optimum mapping limits John Garry via iommu 2022-05-20 23:30 ` Damien Le Moal 2022-05-20 23:30 ` Damien Le Moal via iommu 2022-05-23 6:53 ` John Garry via iommu 2022-05-23 6:53 ` John Garry 2022-05-23 7:33 ` Damien Le Moal 2022-05-23 7:33 ` Damien Le Moal via iommu 2022-05-20 8:23 ` [PATCH 4/4] libata-scsi: Cap ata_device->max_sectors according to shost->max_sectors John Garry 2022-05-20 8:23 ` John Garry via iommu 2022-05-20 23:34 ` Damien Le Moal 2022-05-20 23:34 ` Damien Le Moal via iommu 2022-05-22 13:13 ` [PATCH 0/4] DMA mapping changes for SCSI core Christoph Hellwig 2022-05-22 13:13 ` Christoph Hellwig 2022-05-22 22:22 ` Damien Le Moal 2022-05-22 22:22 ` Damien Le Moal via iommu 2022-05-23 12:00 ` John Garry 2022-05-23 12:00 ` John Garry via iommu 2022-05-24 2:41 ` Martin K. Petersen 2022-05-24 2:41 ` Martin K. Petersen 2022-05-20 21:43 [PATCH 3/4] scsi: core: Cap shost max_sectors according to DMA optimum mapping limits kernel test robot 2022-05-23 11:08 ` Dan Carpenter 2022-05-23 11:08 ` Dan Carpenter 2022-05-23 11:08 ` Dan Carpenter 2022-05-23 11:56 ` John Garry 2022-05-23 11:56 ` John Garry 2022-05-23 11:56 ` John Garry via iommu
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=1653035003-70312-4-git-send-email-john.garry@huawei.com \ --to=john.garry@huawei.com \ --cc=chenxiang66@hisilicon.com \ --cc=damien.lemoal@opensource.wdc.com \ --cc=hch@lst.de \ --cc=iommu@lists.linux-foundation.org \ --cc=jejb@linux.ibm.com \ --cc=joro@8bytes.org \ --cc=linux-doc@vger.kernel.org \ --cc=linux-ide@vger.kernel.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-scsi@vger.kernel.org \ --cc=liyihang6@hisilicon.com \ --cc=m.szyprowski@samsung.com \ --cc=martin.petersen@oracle.com \ --cc=robin.murphy@arm.com \ --cc=thunder.leizhen@huawei.com \ --cc=will@kernel.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.